Skip to main content
Fig. 5 | Microbiome

Fig. 5

From: Microbiota long-term dynamics and prediction of acute graft-versus-host disease in pediatric allogeneic stem cell transplantation

Fig. 5

Machine learning-based prediction of aGvHD severity from the pre-HSCT oral microbiota composition. A Relative abundances the 12 most abundant families over time in the oral cavity in patients with aGvHD grades 0–I versus II–IV. B Importance plot of top 20 predictive oral ASVs identified by the svmLinear model with importance scores indicating the mean decrease in prediction accuracy in case the respective ASV would be excluded from the model. The final cross-validated svmLinear model predicted aGvHD (0–I versus II–IV) from the abundances of oral ASVs pre-HSCT with 92% accuracy (95% CI: 73 to 99%). The ASVs that were also confirmed by Boruta feature selection are indicated with asterisk. C Conditional inference tree (CTREE) displaying ASVs identified as significant split nodes by nonparametric regression for prediction of aGvHD. Numbers along the branches indicate split values of variance stabilized bacterial abundances. The terminal nodes show the proportion of samples originating from patients (n = number of represented samples) with aGvHD grades 0–I vs II–IV. D Boxplots depict the log-transformed relative abundances of the predictive ASVs at time points up to the transplantation in aGvHD grades 0–I compared with grades II–IV patients. E Trajectories of Prevotellaceae and Actinomycetaceae ASVs that were identified by tree-based sparse LDA, including ASV 226 and ASV 568 that were predictive for aGvHD (bold lines), in patients with aGvHD grades 0–I vs II–IV

Back to article page